: Schizophrenic patients have a deficit in inhibitory gating of auditory sensory information. This can be demonstrated using a P50 auditorv evoked potential paradigm. In this paradigm, a schizophrenic patient fails to decrease the amplitude of the second of two closely paired auditory click stimuli. In contrast, normal controls decrease the amplitude of the second sound, compared to the first. This is a failure of inhibitory neuronal processing or of the ability to filter or "gate" incoming auditory stimuli. Treatment with multiple conventional neuroleptics fails to normalize the gating deficit in schizophrenic patients. In contrast, treatment with clozapine. a prototypical atypical antipsychotic, has normalized P50 auditory gating in neuroleptic-resistant schizophrenic patients. This normalization has lasted up to 2 years in some patients. In contrast, treatment with risperidone did not normalize P50 gating in a similar group of patients. Although both antipsychotics have a high 5HT2/D2 affinity, only clozapine blocks the 5HT3 receptor. P50 auditory gating is mediated by the alpha-7 nicotinic receptor, which is deficient in schizophrenic patients. Treatment with nicotine only briefly ameliorates the gating impairment because the alpha-7 nicotinic receptor is a low affinity, rapidly desensitizing receptor. Blockade of the 5HT3 receptor releases acetylcholine in the hippocampus that may help ameliorate the gating impairment. We also hypothesize that a combination of dopaminergic blockade and 5HT3 blockade may be necessary for maximal antipsychotic effect. We will test this hypothesis by studying P50 auditory gating and clinical improvement in 4 groups of 25 neuroleptic-resistant schizophrenic patients, treated with one of the following: clozapine (a 5HT3 antagonist), risperidone (no 5HT3 effect), olanzapine (similar to clozapine; 5HT3 antagonism), and another group treated with haloperidol supplemented by ondansetron (a specific 5HT3 antagonist). In each of the above experiments we will also study Mismatch Negativity (MMN), pre-pulse inhibition of acoustic startle (PPI), P300, and N400. Changes in these measures will be correlated with changes in P50 gating. This RSDA II will allow the P.I. to visit the labs of the researchers who have studied MMN, PPI, P300 and N400 with senior investigators in those measures, and to study clozapine's electrophysiology in multiunit recording experiments with electrodes implanted in superior temporal gyrus, reticular thalamic nucleus, hippocampus, medial septal nucleus, and frontal cortex. He will also experiment with the multivariate statistical techniques needed to analyze multiunit data.